The effect of ecstasy/polydrug use on prospective memory and executive processes

(1)

1

THE EFFECT OF ECSTASY/POLYDRUG USE

ON PROSPECTIVE MEMORY AND

EXECUTIVE PROCESSES

by

Florentia Hadjiefthyvoulou

A thesis submitted in partial fulfilment for the requirements of the degree of Doctor of Philosophy at the University of Central Lancashire

(2)

2 Student Declaration

Concurrent registration for two or more academic awards

I declare that while registered as a candidate for the research degree, I have not been a registered candidate or enrolled student for another award of the University or other academic or professional institution

Material submitted for another award

I declare that no material contained in the thesis has been used in any other submission for an academic award and is solely my own work

(3)

3

Acknowledgments

I would like to thank my parents for all their emotional and financial support all these years. I would also like to thank my friends, and in particular Eirini Tatsi, for their emotional support and understanding. I would also like to thank Dr Catherine A. Montgomery for her valuable help in recruiting participants and allowing me to use the facilities in Liverpool John Moores University. Also, I would like to thank Dr Nikola J. Bridges for all her academic support and guidance not only for the duration of my research degree but also for as long as I have been attending UCLAN. Finally, I would like to express my enormous appreciation and gratitude to Professor John E. Fisk for his constant support, help and mentorship over the last four years. I would like to thank him for all his patience, suggestions on endless drafts and for always steering me in the right direction.

(4)

4

Abstract

The purpose of this thesis was to examine the range of prospective memory and executive function deficits in ecstasy/polydrug users and the role of these processes in accounting for the observed prospective memory performance deficits. Using a variety of laboratory and self-report measures of prospective memory and a self-report measure of executive function, ecstasy/polydrug users were tested in laboratory settings on measures of event and time-based, short and long term prospective memory as well as on a wide range of executive function components. It was found that ecstasy/polydrug users in relation to non-users experience more general prospective memory problems as ecstasy/polydrug-related deficits were evident on both time and event-based and short and long-term prospective memory. Ecstasy/polydrug users also demonstrated deficits on executive processes suggesting that recreational drug users are impaired in a broader range of executive function and ecstasy/polydrug-related deficits are not restricted to the three-model component of executive function. It was also found that executive dysfunction is associated with poorer time-based prospective memory and perhaps some of the drug related prospective memory deficits are mediated by drug related executive function impairment. Finally, although few prospective memory or executive function performance deficits were evident among cannabis-only users a trend was evident in all investigations; ecstasy/polydrug users perform the worst, cannabis-only users at intermediate levels and drug-naïve perform the best. The most striking finding of the present thesis was that the recreational use of cocaine was associated with PM deficits; an association that consistently emerged in all studies of PM performance. The outcomes of the present thesis provide a fruitful direction for future research.

(5)

5 Abbreviations

5-HT 5-Hydroxytryptamine 5-HIAA 5-Hydroxyindoleaceticacid ABI Aquired Brain Injury

ADHD Attention/deficit hyperactive disorder ANOVA Analysis of Variance

ASD Autism Spectrum Disorder BA Broadmann’s area

BD Bipolar Disorder

BRIEF Behavioural Rating Inventory Executive Function CAMPROMPT Cambridge Prospective Memory Test

CANTAB Cambridge Neuropsychological Test Battery CE Central Executive

CFQ Cognitive Failures Questionnaire CSF Cerebrospinal Fluid

DEX Dysexecutive Questionnaire EEG Electroencephalograph EF Executive Function

EMQ Everyday Memory Questionnaire FAB Frontal Assessment Battery

fMRI Functional Magnetic Resonance Imaging HKLLT Hong Kong List Learning Test

MANCOVA Multivariate Analysis of Covariance MANOVA Multivariate Analysis of Variance MCQ Memory Compensation Questionnaire MDA Methylenedioxyamphetamine

MDD Major depressive disorder

MDMA Methylenedioxymethamphetamine OCD Obsessive Compulsive Disorder PASAT Paces Auditory Serial Addiction Task PFC Prefrontal Cortex

PM Prospective Memory

PMQ Prospective Memory Questionnaire

PRMQ Prospective Retrospective Memory Questionnaire PRVP Prospective Remembering Video Procedure PTA Post Traumatic Amnesia

RAVLT Rey Auditory Verbal Learning Test RBMT Rivermead Behavioural Memory Test rCBF Regional Cerebral Blood Flow RM Retrospective Memory

(6)

6 ROCF Rey-Osterrieth Complex Figure

ROI Regions Of Interest

SAS Supervisory Attentional System SERT Serotonin Transporter

SPECT Single Photon Emission Computed Tomography TBI Traumatic Brain Injury

THP Tryptophan Hydroxylase TMT-B Trail Making Test-B TOH Tower Of Hanoi TOL Tower Of London TWTE Test-Wait-Test-Exit

VSWM Visuospatial Working Memory WCST Wisconsin Card Sorting Task WM Working Memory

WMS-R Wechsler Memory Scale-Revised WWW World Wide Web

(7)

7

Table 7.1. Demographical and background drug use variables for users and nonusers………...…………...………134 Table 7.2. Scores on laboratory and self-report measures of real world memory for ecstasy/polydrug users and non-ecstasy users…………...………136 Table 7.3. Correlations between Real World Memory measures and duration of abstinence for the major illicitdrugs……….……….…...142 Table 7.4. Correlations between Real World Memory measures and lifetime use and frequency of use for the major illicitdrugs……….…………...144 Table 7.5. Inter-correlations between the laboratory and self-report measures of Real World Memory……….…..…...………...149 Table 7.6. Inter-correlations between the self-report measures of Real World Memory……….………...……….…..…148 Chapter 8

Table 8.1. Age, intelligence, years of education, cigarette and alcohol use by group………..……….169 Table 8.2. Indicators of illicit drug use….………..………...……….170 Table 8.3. Outcomes for the Prospective Memory, Executive Functions, and memory measures by group………172 Table 8.4. The Relationship between time and event-based Prospective Memory and memory functions………..……..………174 Table 8.5. The Relationship between time and event-based Prospective Memory and indicators of illicit drug use……….……....176 Chapter 9

Table 9.1. Demographical variables for illicit drug users and nonusers……...190 Table 9.2. Background drug use variables for illicit drug users….…………...192 Table 9.3. Performance on the self-report BRIEF-A measure for ecstasy/polydrug, cannabis-only, and nonusers of illicit drugs………..……….194 Table 9.4. Simple correlations and semi-partial correlations (from regression) between BRIEF subscales and aspects of drug use……….196

(12)

12 Chapter 10

Table 10.1. Age, intelligence, years of education, cigarette and alcohol use by group………..……….214 Table 10.2. Indicators of illicit drug use…..………..…216 Table 10.3. Scores on laboratory measures of Prospective Memory for

ecstasy/polydrug, cannabis only and drug naïve………...218 Table 10.4. Simple and partial correlations between aspects of drug use and Prospective Memory measures………..220 Table 10.5. Simple and semi-partial correlations for Executive Function and Prospective Memory measures in ecstasy/polydrug and cannabis only users..222 Table 10.6. Correlations between individual components of Executive Functions and Prospective Memory measures………224

(13)

13

Chapter 1: Thesis Overview

The aim of this thesis is to provide an insight into the impact of ecstasy/polydrug use on aspects of cognition such as Prospective Memory and Executive Functioning that are involved in our everyday activities and play a crucial role in our day-to-day functioning. In Chapter 2 the psychological processes that underline remembering to perform an intended action are discussed. A concise account of the classification of prospective memory, the major theoretical models and the neuroanatomical basis of prospective memory is provided. Chapter 3 evaluates the multidimentional construct of executive processes by exploring the most established executive function models and their biological underpinnings.

Chapter 4 explores the effect of Ecstasy (MDMA) on the brain and its biological underpinnings in both animal and human studies. Chapter 5 provides a concise account of prospective memory deficits in recreational users of ecstasy throughout the literature. It also discusses the different experimental approaches adopted to investigate the effect of ecstasy use on this important aspect of day-to-day memory functioning. Chapter 6 explores the plethora of studies investigating the effect of recreational drug use on the three major components of executive function, updating, shifting and inhibition. It therefore summarises most important findings in this area in order to establish a coherent understanding of the ecstasy-related effect on different components of executive function.

(14)

14 Chapters 7, 8, 9 and 10 are the empirical chapters of this thesis that investigate prospective memory and executive processes in ecstasy/polydrug users. Chapter 7 investigates the impact of ecstasy/polydrug use on real world memory i.e., everyday memory, cognitive failures and prospective memory adopting both laboratory-based and self-report measures of prospective memory. In chapter 8 the range of laboratory measures of prospective memory is augmented by the use of the CAMPROMPT test battery in order to investigate the impact of illicit drug use on event and time-based prospective memory in a sample of cannabis only, ecstasy/polydrug and drug naïve controls. Measures of retrospective memory and learning are also administered in this chapter. Chapter 9 investigates the impact of recreational use of ecstasy on executive processes using a self-report measure of executive function; the Behavioural Regulation Index of Executive Function- Adult Version (BRIEF-A). In chapter 10 both prospective memory and executive function measures are adopted in order to investigate the role of executive processes in accounting for prospective memory deficits observed in ecstasy/polydrug users.

The final chapter is the general discussion of the findings from all four empirical chapters. Consequently, Chapter 11 discusses the findings of this thesis in terms of their implications for recreational drug users, identifies possible limitations and provides directions for future research.

(15)

15

Chapter 2: Real World Memory

and Prospective Memory

Chapter Overview

The purpose of this chapter is to explore the distinct form of memory known as prospective memory and evaluate its impact on everyday functioning. Although the concept of prospective memory has been investigated extensively for the past 30 years, it still remains somehow elusive. Different definitional approaches have been discussed in the literature debating the role of retrospective memory in prospective remembering and the importance of non-cognitive components (such as motivation, reward or conflicting goals) in the successful completion of prospective memory tasks (Einstein & McDaniel, 1996). The different approaches and theoretical models are therefore discussed in this chapter.

When people complain about how poor their memory is, they don’t usually refer to the intricacies in remembering the title of a film they watched days ago or remembering a newspaper article. They usually refer to their everyday cognitive lapses and the failure to recognise acquaintances, forget important events that occurred the previous day, forget the location of familiar objects around the house or forget to take essential objects when leaving the home or office and so on. These aspects of memory lapses fall under the term real world memory and refer to everyday memory (Sunderland et al., 1983) and cognitive failures (Broadbent

(16)

16 ability to remember to attend a meeting, pass on a message or perform everyday intended actions such as remembering to buy milk from the store; an aspect that has been coined as Prospective memory (PM). According to Brandimonte, Einstein and McDaniel (1996) PM refers to the ability to perform activities in the future or simply to “remember to remember”. The focus of memory research was traditionally on the recollection of past events and information or retrospective memory (RM). One of the most important reasons as to why PM has gained increased attention in recent years (Crawford et al., 2003; Kliegel et al., 2000; 2001; 2005; Marsh & Hicks, 1998) is the extent to which PM lapses can interfere with an individual’s everyday functioning. For example, forgetting to buy milk from the store on your way home or forgetting to pick up your dry-cleaning seems inconsequential. Forgetting to take your medication, miss important appointments or interviews, however, can have serious consequences.

Loftus in 1971 was the first researcher to focus on PM. Subsequent research in this new memory field was very slow due to the fact that only a few researchers were interested in this aspect of memory. A milestone for PM was the publication of the first book on the topic in 1996 by Brandimonte, Einstein and McDaniel which although focusing on only the main developments in the area, identified important aspects for future research. From then on, PM has generated considerable interest and has become an important research focus for some researchers (McDaniel & Einstein, 2000; Einstein & McDaniel, 2005)

(17)

17 2.1 Prospective Memory and delayed intentions

According to the literature, PM is the ability to perform activities in the future (Brandimonte, Einstein & McDaniel, 1996; Kliegel et al., 2001; Kliegel et al., 2005) and represents a form of explicit episodic memory that involves the completion of intentions that cannot be realized when initially formed (Ellis, 1996). The ability to retain, recall and realise intentions is an important aspect in everyday memory failures, more specifically in PM (Eldridge, Sellen and Bekeian, 1991; Terry, 1988). Evidence from diary studies suggests that nearly half (West, 1984) or even up to 70% (Terry, 1988) of memory failures in the real world context involve the forgetting of intentions rather than the forgetting of information. Consequently, in order to capture the multidimentional concept of PM, understanding the role of delayed intentions is essential. According to Kvavilashvili and Ellis (1996), delayed intentions are the ones that must be retained and recalled at another moment in the future.

Ellis (1996) distinguishes five phases that are involved in the realization of a delayed intention; Formation and encoding of intention and action (associated with the retention of an action i.e., what you want to do, an intent i.e., the decision to do something and the retrieval context describing the criteria for recall i.e.,

when the intention and action should be retrieved), Retention Interval (refers to

the delay between encoding and the initiation of a potential performance interval),

Performance Interval (refers to the performance interval or period when the

intended action should be retrieved), Initiation and Execution of Intended Action and Evaluation of Outcome.

(18)

18 Ellis (1996) also suggested that for the realization of a delayed intention both prospective and retrospective components are important and that the first phase of the model (formation and encoding of intention and action) forms the retrospective component of the intention and the remaining phases the prospective component. In relation to this, Crawford et al., (2003) argued that PM is concerned with the timing of when things are to be remembered as opposed to RM that is concerned with what should be remembered, and although PM is distinct but not entirely independent of RM, both memory processes are essential to carry out a successful PM task. This chapter, however, will be concentrating on the PM component and its distinct variations.

2.2 Classification of Prospective Memory

As a cognitive construct, PM is more rigorously defined than the typical characterization “remembering to do something in the future” (Marsh & Hicks, 1998). Hereby, in order to capture the many cognitive variables that affect prospective remembering different classes of PM tasks have been proposed through the literature. For example, Kvavilashvili and Ellis (1996) classified PM tasks according to variations in (a) the encoding phase (i.e., importance or pleasantness of task), (b) the retrieval phase (i.e., event- vs time- based tasks), (c) the storage/retention phase (i.e., short- vs long- term delay) and (d) the performance phase (i.e., short or long). Other suggested classes of PM tasks also refer to the complexity of the PM activity (Einstein et al., 1992) and whether the task is habitually or infrequently performed (Harris, 1980). Other important variables affecting prospective remembering include the retrieval context and the

(19)

19 strategies people adopt for remembering (Harris, 1980). Regardless of the importance of these variables the most widely investigated aspect of PM tasks has been the retrieval phase; the focus of the present thesis.

2.3 Retrieval phase

Einstein and McDaniel (1990), proposed that the retrieval phase of PM can be divided into two main classes; time-based PM and event-based PM. Retrieval phase is probably the most researched and debatable phase of prospective remembering and involves the way in which delayed intentions are realized i.e., cued by the monitoring of time or cued by external environmental factors; hence the concept of both time-based tasks and event-based tasks.

2.3.1 Time-based Prospective Memory

The term time-based PM is given for the type of retrieval of a delayed intention, that requires time monitoring i.e., an intention to be performed at a particular time or after a specific amount of time has passed (McDaniel & Einstein, 2000). The best known experiment on time-based PM is the study by Ceci and Bronfenbrenner in 1985 that explored the development of time-based PM in 10 and 14 year old children. In their study, children had to remove cupcakes from the oven after a delay of 30 minutes while they were engaged in a popular video game in a room with a clock for time monitoring. Children had to carry out the task either in a familiar context (their home) or in a laboratory. The authors found that, overall, children checked the clock more often in the laboratory setting and that

(20)

20 the task success was higher in the laboratory than in their familiar context. However, according to the authors, the number of clock checks cannot predict task success as younger children in the familiar context checked the clock more often than older children but success rate was higher in older children. Instead it was the effective and strategic allocation towards the end of the baking period that lead toward the successful completion of the task. Therefore, those children with better PM performance tended to intensify their time monitoring activities more towards the end of the baking period. Consequently, this investigation suggests age-related changes in the development of time-based PM and that strategic clock checking is adaptive and increases PM task success rate.

Craik (1986) went on to suggest that retrieval performance depends on self-initiated or attention demanding processes as opposed to being dependent on environmentally cued automatic processes. Given that the attentional resources essential for processing task relevant information decline with age (Hasher &Zack, 1979), Craik predicted that age-related changes in performance will be larger on PM tasks than on other types of memory processes. Einstein and McDaniel (1990;1996), however, argued that relative to event-based PM, time-based PM performance is more dependent on self-initiated resource demanding processes and therefore that age-related performance would be more pronounced in time-based PM rather than event-based. Although their view was supported by the outcome of some studies (Einstein et al., 1995; Park et al., 1997) a number of other investigations have shown that older adults perform better than younger adults on time-based tasks given that assessment occurs in the context of their everyday life (Martin, 1986; Maylor, 1990; Rendell & Craik, 2000; West, 1988).

(21)

21 By way of contrast, Birt (2001), in a meta-analysis showed larger age effects on time-based task than on event-based tasks. It is difficult to draw definitive conclusions from these studies as the literature is somewhat contradictory as to whether age-related differences are more common in event- or time-based tasks.

Context importance

Confusion in the literature might be attributed to the fact that the context in which memory performance is assessed is largely ignored. Hereby, in Birt’s meta-analysis, it was found that the age effect is larger for time-based rather than event-based tasks; a result that is consistent with Einstein and McDaniel’s (1990) prediction. In relation to this, studies on aging have shown that time-based PM performance is affected because self-initiated processing is impaired in older adults whereas event-based task performance is not affected (Einstein et al., 1995; Katai et al., 2003; Kliegel et al., 2001; Khan et al., 2007). Another interesting finding from Birt’s meta-analysis was that naturalistic studies (i.e., in the context of a familiar setting and everyday life) showed a reverse age effect suggesting that older adults are more successful than young adults on time-based tasks when those tasks are performed in the context of their everyday life.

In order to understand the importance of the context in which memory performance is assessed, Einstein and McDaniel (1990;1996) have pioneered two computer based paradigms (one event-based and one time-based) to mimic real-life prospective remembering when people are busily engaged in other activities. Accordingly, in their time-based laboratory paradigm participants monitor a clock

(22)

22 and respond at fixed intervals while completing an attention demanding task. Much of the literature in this area uses these paradigms to assess time-based and event-based prospective remembering in a laboratory environment.

Task importance

Another factor affecting the performance of PM, in particular time-based PM, is the perceived importance of the delayed intention. Some studies suggest that the importance of the task can affect performance and therefore the successful completion of the delayed intention (Ellis, 1988; Kvavilashivili, 1987; Meacham & Singer, 1977; Kliegel et al, 2001). Diary studies have reported that successful remembering was higher for important appointments (Andrzejewski et al., 1991) and that there was a positive relationship between recollection of the intention and the perceived importance of the intention (Ellis,1988). Meacham and Singer (1977) also suggested that high-incentive is predictive of better performance. Participants who were given a monetary incentive to return four prepaid postcards on specified dates performed better than participants with no incentive. Similarly, Kvavilashvili (1987) demonstrated a significant positive effect of task importance on PM performance. As a way of contrast, Goschke and Kuhl (1993) reported that subjective importance of delayed intentions had no effect on their recall. Kliegel

et al. (2000), in a series of experiments labelled a time-based PM task as

important as opposed to a cover task. Their findings suggested that the importance of the task leads to a selective increase of attention allocation towards the PM task, particularly during the last period before the completion of the task. They also suggested that the accuracy of prospective remembering can be influenced by

(23)

23 affecting attention allocation at specific phases of the process when the importance of the task is manipulated.

In relation to this, Kliegel et al. (2001), in a second experiment investigated the assumption that event-based PM is an automatic process and does not rely on attentional resources. It was found that PM performance was unaffected even with an addition of a task that increased overall demands of the ongoing activities. In addition to this they found that at least some event-based PM tasks are mediated by relatively automatic processes and require very little attention for successful performance. In terms of task importance, the authors found that importance has an effect on the time-based but not event-based PM tasks. Furthermore, the importance of the task improved PM to the degree the task requires the strategic allocation of attentional resources.

2.3.1.1 Theoretical models of time-based prospective memory

According to Coren and Ward (1989), attentional resources support the process of monitoring. Humans have a limited attentional capacity so higher cognitive load can negatively affect monitoring of time in time-based tasks. To investigate this assumption, Khan, Sharma and Dixit (2008) explored the relationship between cognitive load and event- and time-based PM. They found that performance deteriorated in both PM tasks as the cognitive load increased. Nevertheless, performance under an event-based task showed less error compared to the time-based task suggesting that monitoring is more crucial for time-time-based PM.

(24)

24 Although it is clear that retrieval in time-based PM is fundamentally different from that in event-based (Einstein & McDaniel 1996), there are only a few empirical studies that examined the nature of retrieval in time-based tasks (Ceci &Bronfenbrenner, 1985; Cicogna et al., 2005; Cook et al., 2005; Einstein et al., 1995; Park et al., 1997; Redell & Craik, 2000). One finding that emerged from the existing literature involves the participant’s time monitoring behaviour prior to the critical time to remember. More specifically, it has been suggested that the frequency of rehearsal in a time-based task is positively correlated with PM performance (Harris & Wilkins, 1982; Einstein et al., 1995; Park et al., 1997). Prior to the emphasis on attentional resources, in order to explain these findings an early theoretical account was offered by Harris and Wilkins (1982): the

Test-Wait-Test-Exit (TWTE) model of time-based PM.

This model proposes that people encode a future task and then wait until the time is appropriate to carry out the intended task. For example, a person needs to ‘test and wait’ until the time is appropriate to take the cookies out of the oven before they burn. When the action is carried out, then the ‘test and wait’ cycle is stopped (‘exit’). Consequently, successful performance is dependent on monitoring the time during the critical period. This contrast with the more recent perspectives on time-based PM, as described in later studies (Einstein et al., 1995; Park et al., 1997) in which the time monitoring aspect is a self-initiated process that requires attentional resources.

Despite the insights provided by these two models the main question remains: What is the nature of these self-initiated processes? Harris and Wilkins suggested

(25)

25 that the intention spontaneously pops into a person’s mind or is triggered by some incidental cues in the environment. An experiment, where time references in a film made participants aware of their PM tasks, supports another theoretical model of time-based prospective memory proposed by Wilkins and Baddeley (1978), the “random walk” model. This model underlines the significance of incidental external or internal cues in remembering intentions. In contrast to this model Wilkins and Baddeley proposed that our mind is a multidimentional semantic space and a trace is formed in this space when an intention is formulated. Our thoughts, however, do not remain in this space throughout the delayed interval; instead they move in various parts of this area randomly and depend on the stimuli we come across in the environment and the activities that we are engaged in. If near the time of the execution of the intention, those thoughts accidentally wander around the trace of the intention then it is likely that we realise that we should carry out an intention, thus successfully carrying out the PM task. This model does not therefore regard the retrieval of the intention as a self-initiated process. Instead, it proposes that the timely remembering of intentions depends entirely on incidental factors.

In order to investigate which of the two models is correct and therefore determine what brings a time-based PM task into our minds during the retention interval, Kvavilashvili and Fisher (2007) conducted a series of three studies. The authors first investigated self-report rehearsal processes in naturalistic time-based PM tasks and compared them with event-based PM tasks. The participants were expected to phone the experimenter at a prearranged time (time-based) or after a text message (event-based). Participants also recorded the details of occasions

(26)

26 when they thought about the intention during a seven day delay interval (long delay interval). It was found that the intention is either triggered by incidental cues or periodically pops into one’s mind for no apparent reason. This opposes previous literature and suggests that rehearsal and retrieval of time-based PM could be a more automatic process than previously thought.

Kvavilashvili and Fisher (2007) also emphasized the importance of the storage/retention phase in the successful completion of a time-based PM task. They suggested that the process of retrieval (i.e., automatic/self-initiated) can depend on storage/retention phase. For example, most laboratory studies have used short time delays (Sellen et al., 1997), so participants are likely to keep the task in mind for the entire delay period. Therefore retrieval processes in short-term laboratory tasks can be regarded as self-initiated and deliberated. On the other hand, remembering time-based PM tasks in everyday life with long delay intervals cannot occur in the same way, as people are engaged in more activities during the delay period. In addition, self-initiated rehearsals, occurring when people are engaged in planning their daily activities, were reported in few cases regardless of age, and were lower than rehearsals triggered by incidental cues. These results suggest that a great variety of cues, such as internal or external incidental cues or cues completely unrelated to the intention, can act as triggers and promote successful prospective remembering.

These findings are therefore more in line to Harris and Wilkins (1982) model of TWTE. Kvavilashvili and Fisher (2007) also suggest that retrieval of event- and time-based PM is not mediated by fundamentally different processes and that

(27)

27 thoughts about the task/intention occur via three different routes in both event-based and time-event-based PM: i.e., by rehearsal prompted by incidental external or internal cues, by self-initiated planning thoughts or by no apparent triggers. This suggests that the difference between event- and time-based PM is more quantitative than qualitative since representations of event-based tasks have low level of activation that is sufficient to trigger remembering with the occurrence of a target and/or event in the environment. On the contrary, the activation levels in time-based tasks may be greater and fluctuate over time leading in periodic conscious thoughts about the task.

2.3.2 Event-based Prospective Memory

Beyond the context of time-based PM, event-based PM has been studied extensively in its own right. The term event-based PM tasks refers to the situation where the intended action is performed at the occurrence of an external or environmental cue or event (Einstein & McDaniel, 1990). The authors in their laboratory paradigm of event-based PM, tried to mimic real life event cued prospective remembering by giving their participants one or two words to remember and instructing them to press a key whenever the target words appeared while they were busily involved in an ongoing task. Although the laboratory paradigms of event-based and time-based PM are very useful in assessing PM under laboratory condition in a naturalistic manner, they do not completely capture more complex PM situations in which several delayed actions are planned to be executed (Kliegel et al., 2000).

(28)

28 In relation to this Ellis (1996) suggested that the more complex situations are likely to include planning processes such as forming a daily plan of activity. Laboratory tasks, however, do not involve such planning. In an attempt to investigate the potential role of the complexity of processes involved in many PM activities such as developing a plan, remembering the plan and remembering to execute the plan sometime in the future, Kliegel, McDaniel, and Einstein (2000), used a six element laboratory task based on a paradigm by Shallice and Burgess (1991). In this experiment participants had to work within constraints on six subtasks to maximise their total points. This paradigm required the participants to engage with a range of processes that included making a plan, retaining the intended plan and executing a series of multiple intentions. The PM component of this task was that participants had to initiate the six subtasks on their own at a specific point during the test. It was found that the planning and executing of PM tasks have to be distinguished, since formulating a plan did not overlap substantially with the manner in which PM tasks were executed.

The involvement of Retrospective Memory (RM) in the execution of PM tasks needs to be addressed when performing a PM task as the literature suggests that both the prospective elements and RM are crucial for successful prospective remembering (Crawford et al., 2005). Kvavilashvili (1987) found that remembering an intention at an appropriate moment and remembering content or facts acquired in the past might be considered as two separate forms of memory, suggesting that RM and PM are somehow different.

(29)

29

Retrieval context and attentional resources in event-based Prospective Memory

According to the literature, the retrieval context plays a key role in prospective remembering. McDaniel and Einstein (1992) argue that distinctive cue words, as opposed to background words, can increase PM performance. For example, more specific cues such as the word “tiger” can trigger better performance than more general cues such as “animal”, given that the cue is relevant to the task (Einstein

et al., 1995). An important question when investigating cues is whether noticing a

cue is an automatic response or whether conscious attentional resources are essential (Marsh & Hicks, 1998). These possibilities have been associated with a lot of conflicting evidence throughout the literature and two theories have been developed to explain this type of PM retrieval. The first theory assumes that attentional and/or working memory resources need to be allocated to monitoring the environment for the occurrence of the target event (Smith, 2003). Consequently, in order to successfully retrieve an intention, strategic, resource-demanding processes must be employed before the occurrence of the target event. The second theory supports a multiprocess model of PM retrieval that involves several processes (McDaniel & Einstein, 2000). The different approaches to understand the processes involved in event-based PM are discussed below.

2.3.2.1 Theoretical Models of event-based PM

Similarly to time-based PM, a number of theoretical models have been proposed to understand the underlying mechanisms that lead to successful event-based retrieval. According to Guynn (2003), in laboratory PM tasks, the monitoring

(30)

30 process involves a recognition check to evaluate whether the cue presented is the correct one for performing the intended action. If the recognition check indicates that the cue represents a target event then the intended action is executed. Failure to carry out the intention is therefore, due to the person’s failure to initiate a recognition check (in other words failure to monitor) or due to the failure of the recognition check to identify the event as a target. This theory is therefore based on two main assumptions; that monitoring processes require capacity demanding attentional processes and that monitoring processes are essential for prospective remembering to occur. If this is the case then the resource demanding processes required for PM will reduce the attentional resources available for performing ongoing activity and consequently lower the performance success of the ongoing task. This assumption is supported by a number of studies (Cohen et al., 2008; Einstein et al., 2005; Marsh et al., 2003; Smith, 2003). A specific mechanism proposed to support monitoring is the supervisory attentional system (SAS; Shallice & Burgess, 1991) which monitors for a cue signalling the appropriateness of executing the intended action. When a cue is detected the SAS switches attention to the intended action. This suggests that the realisation of an intended action is an attentional process supported by executive attentional systems and not memory processes per se.

By way of contrast, McDaniel and Einstein (2000) proposed a different multiprocess theory suggesting that because of the PM demands in everyday life it is adaptive to have a cognitive system to aid PM retrieval through several processes. So, in addition to the resource demanding processes such as monitoring, prospective remembering can sometimes be spontaneously elicited by

(31)

31 features of the target cue even without resources dedicated to the intention. McDaniel and Einstein (2004) maintain that this spontaneous retrieval in event-based PM can involve a number of processes such as the reflexive-associative

hypothesis, in which the cue is strongly associated with the intention during

planning and the intention is performed reflexively.

In relation to this, Guynn, McDaniel and Einstein (2001) proposed an alternative to conscious cue-focused account based on a memory model proposed by Moscovitch (1994); an “automatic-associative” memory system that consciously attends to external cues which in turn interact with memory traces previously associated with those cues. If there is enough interaction between the external cue and a memory trace then the system delivers awareness of the information associated with the cue, thus mediating PM retrieval. As opposed to cue focus theory, the target event is not necessarily recognised as a cue; it simply stimulates a reflexive associative process bringing the intended action into awareness. The entire pattern implicates both cue-focused and reflexive associative process and more generally supports a multidimensional framework of PM (McDaniel and Einstein, 2000).

As previously discussed, some researchers argue that PM declines with age and a number of experiments using event-based PM to appear support this assertion (see review Henry et al., 2004). However, other studies report no age differences in event-based PM (Einstein and McDaniel, 1990). In an attempt to understand this anomalous pattern, Einstein and McDaniel (2005) used the multiprocess point of view and suggested that age differences depend on whether the PM task uses focal

(32)

32 or non-focal target events. According to Einstein and McDaniel (2005) a focal PM cue is the one that stimulates the spontaneous retrieval of an intention without the need to employ strategic monitoring processes. With non-focal targets attention-demanding processes (i.e., monitoring) are essential for prospective remembering and according to Craik (1986) these resources decline with age. Conversely, focal targets require spontaneous retrieval which is assumed to stay relatively intact with age. To support this assumption, Rendell and Craik (2000) found minimal age-related declines in event-based PM when the event was focal. In contrast, when the event was non-focal the age differences were more pronounced.

According to McDaniel, Guynn, Einsten and Breneiser (2004), spontaneous retrieval, as opposed to monitoring, can occur even when no resources are devoted to monitoring for the target during or prior to the occurrence of the target. To support this, Einstein et al.’s (1995) results from a study comparing performance on event-based PM tasks between older and younger adults suggest a large automatic component to event-based PM. Marsh and Hicks (1998) suggested that these mixed findings can be explained by the character of the demands that the tasks place on working memory and that poorer event-based PM performance depends on an attention demanding component and therefore might be correlated with measures of central executive functioning.

To support this view, the notice-search model (Kliegel et al., 2001; Logie et al., 2004) has also been proposed. This model suggests that for successful PM, familiarity and probe search are required. When people encounter the PM cue they get a sense of familiarity (noticing) which may then prompt a more conscious

(33)

33 probe of memory (search) to determine what the cue means. Therefore, there are two stages in a successful event-based PM task: the stage of noticing or a feeling of familiarity and the search stage. Burgess (2000b) suggested that PM task completion requires many of the skills that are commonly described as executive processes. Successful completion of intentions rely on the operation of a number of different cognitive processes including attention, action control and memory (Dobbs & Reeves, 1996; Ellis, 1996). In particular the literature on PM addresses an important debate on the attentional or strategic demands of PM task retrieval evaluating the notice-search (strategic component) and automatic activation models. According to West and Craik (1999) older adults are more prone to lapses of intention and are believed to suffer from attentional or executive deficits. These failures are associated with changes in neural activity in a region thought to be responsible for the implementation of cognitive control. It is therefore reasonable in order to further understand the underlying mechanisms of PM to look at changes in neural activities during the realisation of PM tasks.

2.4 Neuroanatomical basis of event-based and time-based PM

Many investigations of PM implicate the role of the frontal lobes, more specifically the involvement of the prefrontal cortex (PFC), in the realisation of delayed intentions. Processes in both event- and time-based PM can be linked with frontal lobe activity. This evidence is coming from patients with frontal lobe dysfunction (Fuster, 1997) and age-related literature (McFarland & Glisky, 2009). Although the literature has been somehow elusive as to whether age is responsible for greater decline in time-based or event-based PM, a vast body of research

(34)

34 agrees that younger adults perform better than older adults in PM tasks that require self-initiated processing; i.e., time-based PM tasks (Einstein et al., 1995; Einstein et al., 1997; Maylor, 1996; McDaniel et al., 2004; Park et al., 1997). Time-based PM tasks, although requiring many of the same processes of event-based PM, have greater monitoring demands and are more likely to be entirely self-initiated (Craik, 1986; Einstein and McDaniel, 1990; 1996).

As discussed previously, time-based PM tasks require the formation of an association between cue and intention, the maintenance of this intention over a delayed period, the division of attention between tasks, monitoring the environment for a cue and the interruption and inhibition of ongoing activities. Fuster (1997) showed that these operations are impaired in frontal lobe patients; thus implicating the role of the frontal lobe in time-based PM tasks. Age-related declines in frontal lobe functions have also been showed by West (1996). Support for age-related declines comes from a range of studies. For example, morphological evidence shows disproportional volume loss in the PFC in relation to other brain areas in older adults (Raz et al., 2005).

Neuroimaging studies also suggest that the anterior PFC and more specifically Broadmann’s area10 (BA10) is likely to be of central importance to PM (Okuda et

al., 1998; Burgess et al., 2001,2003; Simons et al., 2006). In particular, Okuda et al. (1998) employing Positron Emission Tomography (PET) examined the

functional neuroanatomy of PM by examining changes in regional cerebral blood flow (rCBF). They found increased activity in the left frontal pole, the ventrolateral PFC (BA 8/9/47) and anterior cingulate (BA24) during a PM task.

(35)

35 Burgess, Quayle and Frith (2001) also found increased activation in BA10 (bilaterally) across several cognitive tasks. In their study activation during an ongoing task was compared to activation in two PM conditions (i.e., cue identification and intention retrieval). Increased activation relative to a control task in bilateral frontal pole, right lateral, prefrontal and parietal cortex was observed. The same authors in a later study (2003) extended their previous findings by showing that this bilateral activation of lateral BA10 that is associated with retrieving a delayed intention was accompanied by a deactivation in medial BA10. In relation to this, an activation was observed in lateral BA10, lateral parietal cortex and precuneus. Den Ouden et al. (2005) found that these increased activations were associated with holding an intention during an ongoing task.

Furthermore, Simons, Scholvinck, Gilbert, Frith and Burgess (2006) measured brain activity (using functional magnetic resonance imaging [fMRI], and a combination of two different PM tasks: words and shapes) while manipulating the demands on either recognizing the appropriate context to act (cue identification) or remembering the action to be performed (intention retrieval). A consistent pattern of hemodynamic changes was found in both PM conditions in anterior prefrontal cortex (BA10), with lateral BA10 activation accompanied by medial BA10 deactivation. These effects were more pronounced when demands on intention retrieval were high. This is consistent with the hypothesis that anterior prefrontal cortex (area 10) supports the biasing of attention between external events (e.g., identifying the cue amongst distracting stimuli) and internal thought processes (i.e., maintaining the intention and remembering the intended actions). These results suggest that whilst cue identification and intention retrieval may be

(36)

36 behaviourally separable, they share at least some common neural basis in anterior prefrontal cortex. PM related activation was also evident in areas outside anterior PFC region such as lateral PFC and parietal cortex. The anterior cingulate cortex was also activated to a greater extent in a cue identification PM task and the posterior cingulate and precuneus showed greater activation in the intention retrieval task (Simon et al 2006; Okuda et al 1998; Burgess et al 2001).

Further evidence for the involvement of the PFC in PM comes from Okuda et al (2002) who looked at PET activation during a time-based PM task. Participants had to clasp their hands either at a time point (time-based) or after a cue (event-based) while performing a mental arithmetic task. Both conditions increased rCBF in frontal and medial temporal regions. The authors however did not compare brain activity between the two tasks or examine decreases in rCBF thus it was unclear if the two tasks made differential demands upon rostral prefrontal brain activity consistent with the age-related literature. Reanalysing Okuda’s et al’s 2002 data Okuda et al., (2007) observed significant rCBF increases in the left superior frontal gyrus (including lateral BA10) for the time-based PM relative to the event-based task. Deactivations within rostral PFC were evident in the medial BA10 as rCBF decreased during the event-based PM task in comparison to the ongoing activity alone. The authors also found that the decrease in medial BA10 during time-based PM was not as significant as in the event-based PM suggesting that deactivation in medial BA10 during PM task are specific to event-based PM. Okuda et al. (2007) also found that during time-based PM the right superior frontal gyrus, anterior medial frontal lobe and anterior cingulate gyrus were more active and that the left superior frontal gyrus was more active in the event-based

(37)

37 condition. The results suggest the involvement of multiple brain regions of rostral prefrontal cortex in both time- and event-based PM.

To conclude, there is growing evidence that the frontal lobes and more specifically the PFC are not the only brain regions that are involved in the realisation of delayed intentions. Regions such as the right dorsolateral and ventrolateral prefrontal cortices, the left frontal pole and medial frontal regions and the left parahippocampal region (Okuda et al., 1998) provide the neuroanatomical basis for PM.

2.5 Chapter summary

The literature on PM has yet to reach a consensus and there are many theoretical models that have been proposed to identify the underlying mechanisms through which intentions in either time-based or event-based actions are retrieved and the factors affecting this retrieval. What is undisputable is the important role of PM in our everyday environment and the need for more investigation in the area. All in all, having discussed the most established theoretical models of both event- and time-based PM and the role of the frontal lobes in the execution of PM tasks, it can be tentatively concluded that time-based PM tasks are reliant on self-initiated processes whereas event-based tasks are considered to be dependent on more automatic processes. On the whole, the research literature has suggested that executive processes such as planning, monitoring or attention are essential for PM performance. It is therefore reasonable to assume that there is an association between executive processes and prospective remembering.

(38)

38 This assumption can be supported by evidence suggesting that PM processes such as dividing attention, monitoring the environment for a cue, associating a cue for intention and interrupting an ongoing activity may also involve planning which is thought to depend on the frontal lobes (Lezak 1982; Shallice, 1982). PM as it has been previously discussed depends on self-initiated and attention demanding resources and therefore PM performance can be correlated with measures of central executive functioning (Marsh and Hicks, 1998). In relation to this, Martin, Kliegel and McDaniel (2003) found that executive processes in older adults were significantly correlated with performance on three PM tasks. To support this view, additional studies have implicated the role of executive processes in PM performance (Kliegel et al., 2000; Kliegel et al., 2008).

More direct evidence of the involvement of executive processes in PM comes from neuroimaging studies that suggest regions of the frontal lobe (such as rostral prefrontal cortex) are involved in supporting both event-based and time-based PM tasks (Burgess et al., 2003; Martin et al., 2003; Okuda et al., 2007; Simons et al., 2006). According to Okuda et al. (2007) these regions are involved in the attentional and executive control aspects of PM functions. Having said that, it will be sensible to evaluate the term executive function and, essentially, its involvement to prospective memory processes. Consequently, the next chapter will look in depth the central executive system and the possible involvement of it in PM performance.

(39)

39

Chapter 3: Executive Function

Chapter overview

Over the last few decades the term executive function has received increased attention. Early models of executive function were restricted to cognitive abilities using a unitary framework while specific components of executive function were not identified and the biological basis of this term was limited to frontal lobes. Nowadays, executive functions are known to represent a rather complex, interrelated set of cognitive abilities critical for adaptive function. Despite the plethora of research and speculation concerning executive functioning the term itself and the conceptualization of it still remains somewhat elusive. The purpose of this chapter is to provide an up to date perspective of executive processes by exploring the most established executive function models and their biological underpinnings.

3.1 What is Executive Function (EF)?

Before discussing the theoretical models which may be found within the literature on EF, it is essential to define the concept. Many definitions have been proposed through the years by different researchers that have influenced research and clinical practices. For example, Lezak (1995) defines EF as a group of superior abilities of organisation and integration; such as anticipating and establishing goals, designing plans and programs, self-regulation and monitoring of tasks.

(40)

40 Similarly, Welsh and Pennington (1988) suggest that EF is “the ability to maintain an appropriate-solving set for attainment of a future goal” (pp. 201). However, according to Gioia, Isquith, Guy and Kenealy (2000), EF is not restricted to cognitive processes but is also characterised by emotional responses and behavioural actions; something that these constructs fail to capture. EF is therefore better described as a collection of interrelated tasks or processes that are responsible for goal-directed or future-orientated behaviour with the executive system acting as the “conductor” that controls, organises and directs cognitive ability, emotional responses and behaviour (Gioia et al., 2001). Having said that, Gioia et al. (2000), identified the key elements of EF that include the anticipation and deployment of attention, impulse control and self-regulation, initiation of activity, working memory, mental flexibility and utilisation of feedback, planning ability and organization and selection of efficient problem-solving strategies.

3.2 Theoretical models of Executive Function

In order to understand the critical role of executive functioning in our everyday lives, researchers throughout the years have tried to provide a theoretical framework of this complex term in order to comprehend how executive dysfunction affects our everyday life and determine the different neural pathways underpinning EF. Although a number of theoretical models of EF have been proposed, no one model has been uniformly accepted. Early attempts to conceptualise EF resulted in unitary models such as Baddeley’s (1986) “Working memory” model or Norman and Shallice’s (1986) “supervisory acting system”. However, later research demonstrated that the unitary view is too simplistic and

(41)

41 that the term EF is more likely to be composed of distinct but interrelated components (Baddeley, 2000; Miyake et al., 2000). Findings that frontal lobe patients rarely exhibit global executive dysfunction provide evidence for fractionation of EF (Bigler, 1988; Pennigton & Ozonoff, 1996). In light of this new evidence, concepts such as the central executive have been modified in an attempt to fractionate the overall construct to derive subcomponents constituting the various control systems. Before discussing fractionated accounts of executive functioning a description of Baddeley’s working memory model will be outlined (Baddeley, 2000).

3.2.1 Working memory model

Baddeley’s model proposes that working memory plays a key role in complex activities and is consisted of four major components; the central executive system, the phonological loop, the visuospatial sketchpad and the episodic buffer. Baddeley (2000) defines working memory as “a limited capacity system allowing the temporary storage and manipulation of information necessary for such complex processes as comprehension, learning and reasoning” (pp. 418). Figure 1 represents Baddeley’s (2000) Working memory model.

(42)

42

Figure 1. Working Memory model (Baddeley, 2000).

According to this model, working memory consists of the “central executive” system that is a limited capacity attentional system and two slave systems; the phonological loop and the visuo-spatial sketch pad (See Figure 1). The shaded areas represent 'crystallized' cognitive systems that are responsible for accumulating long-term knowledge (e.g., language and semantic knowledge), whereas unshaded systems represent 'fluid' capacities (e.g., attention and temporary storage) and are unchanged by learning, other than indirectly via the crystallized systems (Cattell, 1963). The episodic buffer according to Baddeley (2000) is a third slave system that links information across domains to form visual, spatial and verbal information with chronological order (i.e., memory of a story). The episodic buffer is also speculated to have links with long-term memory.

According to Baddeley (1996, 2000), the “central executive” has four main functions. Firstly, it is responsible for selective attention in that it selectively

(43)

43 evaluates a relevant piece of information while ignoring irrelevant information and distractions. Impairment of the central executive therefore results in the failure to evaluate targeted events/stimuli and maintain goal-directed behaviour as the actions of the central executive are influenced by distractions or irrelevant information. Secondly, the central executive is capable of coordinating two or more simultaneous activities by managing sufficient working memory resources across the tasks. The third function of the central executive is the ability to switch attention and respond to a task or situation that requires mental flexibility prevailing habitual or stereotyped behaviours. Impairment of this can result in rigid performance and perseverative behaviour. Finally, the central executive is responsible for retrieving information from long-term memory a crucial function for responding to the demands of the environment.

The Phonological Loop has the ability to temporarily maintain and manipulate speech based information. It is therefore responsible for retaining verbal and acoustic information using a temporary store and an articulatory rehearsal system. Visuo-spatial sketch pad on the other hand, is responsible for holding and manipulating visuospatial information while the episodic buffer is controlled by the central executive and provides space for temporary storage of information. It also has the ability to integrate information from the two slave systems and long-term memory to create a unitary episodic event (Baddeley, 2000).

While Baddeley’s model specifies distinct functions for the central executive, it is unclear as to whether these are performed by a single unitary system or by a collection of discrete and separable executive resources. Furthermore, although

(44)

44 the working memory construct has been studied extensively and is considered to be a well validated model which offers a coherent conceptual framework for describing executive processes and although it accounts for specific patterns of executive impairments, nonetheless it neglects important elements of executive functioning such as goal setting, reasoning and planning.

3.2.2 Miyake et al.’s model of Executive Function

Another theoretical model of EF that has received increased attention is Miyake et

al.’s (2000) model which proposes that the central executive is fractionated with

three components performing separate tasks with varying degrees of competence. Miyake et al. proposed the separability of three executive functions: shifting, updating and inhibition and their contribution to higher level complex executive tasks. The authors focused on these three executive components not only because they have been widely discussed in the literature and there are a number of well-studied cognitive tasks (such as Wisconsin Sorting Card Task (WCST) and Tower of Hanoi (TOH)) that tap each target function, but also because these three components are likely to be implicated in the performance of complex executive tasks. The first component of this model has been proposed as being crucial for understanding the failures of cognitive control in brain-damaged patients and laboratory tasks where the participant is required to shift between tasks. In other words, ‘shifting’, is responsible for shifting back and forth between several tasks or mental sets (Monsell, 1996) and is considered to be an important aspect of executive control (Norman and Shallice, 1986).

The effect of ecstasy/polydrug use on prospective memory and executive processes